In many businesses, data libraries can be quite large, constituting a significant portion of a company's computing service budget and effort. In addition, significant portions of a data library may be archived, often in a way that is quite inefficient when individual data files need to be retrieved. For example, related data files or portions of data files (e.g., those created or accessed on the same day) may be stored on separate storage devices (e.g., optical drives, tapes, etc.). Thus, when a data file or a group of related data files must be retrieved from the data library, it may be necessary to access and/or restore the contents of multiple storage devices.
Even on a single storage device, related data files may be stored at some distance from each other. This is particularly problematic for serially-accessed media, such as tape, since the entire length of the tape may have to be traveled in order to retrieve disparately stored data files. As noted above, if the needed data files are stored on multiple storage devices or on archived storage devices, restoring these data files can be quite labor-intensive and costly.
One reason that storage on tapes or similar storage devices becomes inefficient is that files archived onto a tape expire at different times. As a result, tapes are only partially used and not reused until all files have expired.
Accordingly, there exists a need in the art to overcome the deficiencies and limitations described hereinabove.